Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Sep 23.
Published in final edited form as: Psychol Aging. 2010 Dec;25(4):858–866. doi: 10.1037/a0019622

Psychosocial Predictors of Changing Sleep Patterns in Aging Women: A Multiple Pathway Approach

Cynthia H Phelan 1, Gayle D Love 2, Carol Ryff 3, Roger L Brown 4, Susan M Heidrich 5
PMCID: PMC3780608  NIHMSID: NIHMS245320  PMID: 20731498

Abstract

The purpose of this investigation was to examine changes in the sleep quality of older women over time and to determine whether dimensions of psychological well-being, health (subjective health and number of illnesses), and psychological distress (depression and anxiety) predict these changes. A secondary analysis was conducted using a longitudinal sample of aging women (Kwan, Love, Ryff, & Essex, 2003). Of 518 community-dwelling elderly women in the parent study, 115 women (baseline M age = 67, SD = 7.18) with data at baseline, 8 years, and 10 years were used for this investigation. Participants completed self-administered questionnaires and participated in in-home interviews and observations. Growth curve modeling (GCM) was used to examine the overall linear trajectories of sleep quality. Growth mixture modeling (GMM) was used to examine whether there were different patterns of change in sleep quality over time and to examine baseline predictors of each pattern. Sleep quality declined over time but not for all women. Two distinctly different sleep patterns emerged: good but declining sleep quality (GS) and disrupted sleep (DS) quality. Higher psychological well-being (positive relations with others, environmental mastery, personal growth, purpose in life, and self-acceptance), fewer illnesses, and lower depression scores at baseline predicted reduced odds for membership in the DS group. Future research is needed to examine whether interventions focused on maintaining or enhancing psychological well-being could minimize later life declines in sleep quality.

Keywords: Sleep, Older Women, Psychological well-being, Depression, Health

Declines in sleep quality are a significant problem in community-dwelling older adults, particularly women. The study of women is important because women report higher rates of sleep problems, have a wider variety of sleep complaints, and are more likely to report multiple sleep complaints than men (Foley et al., 1995; Maggi et al., 1998; Middlekoop et al., 1996). The aim of this investigation was to examine the relationship between sleep quality and health (physical and psychological) in older women and to identify factors that might predict changes in sleep quality with aging. Unique to this investigation is whether positive psychological functioning, conceptualized as psychological well-being (Ryff, 1989) in this study, may have an impact on sleep quality in aging. According to Ryff (Ryff, Singer, & Love, 2004) there is an interplay between biopsychosocial processes such that positive psychological well-being will be accompanied by positively functioning physiologic processes. This approach may help explain why some older adults experience less disrupted sleep than others.

The most common sleep disorders in older adults include: insomnia, sleep-related breathing disorders and limb movement disorders (Ancoli-Israel, 2009). The study of sleep has included both subjective (self-report) and objective (polysomnography) measures. The gold standard for the diagnosis and management of sleep-related breathing disorders and limb movement disorders is polysomnography (PSG). However, PSG has drawbacks and is not recommended for use in all sleep investigations. For example, underlying etiologies for insomnia may not be detectable with PSG and consequently PSG is not recommended in the routine diagnosis and treatment of insomnia (Littner et al., 2003). There are limitations in the use of self-report methods as well. Individuals with significant sleep disorders may experience little or no night time distress. Also, evidence suggests that some individuals have sleep state misperceptions and use of self-report measures tend to underestimate sleep time and over estimate disruptions in sleep. Sleep complaints are not sleep disorders in and of themselves. The intent is this investigation is not to diagnose sleep disorders but to describe changes in sleep quality. As such, self-report methods were used.

In the National Institutes on Aging’s multicenter study, Established Populations for Epidemiologic Studies of the Elderly (EPESE), over 50% of older adults reported at least one of four sleep complaints “most of the time” (as cited in Foley et al.,1995). These complaints include (a) difficulty initiating sleep, (b) difficulty maintaining sleep, (c) waking too early, or (d) waking not feeling restored. In a longitudinal study of 1,050 community-dwelling older adults, 37% reported difficulty falling asleep, 29% reported fragmented sleep, and 19% reported early rising. Two years later the percentages increased substantially (75%, 69%, and 47% respectively (Ganguli, Reynolds, & Gilby, 1996).

Studies of changes in sleep physiology using PSG noted a reduction in total sleep time and sleep efficiency (i.e., ratio of time asleep to time in bed), lighter sleep, and increases in nighttime awakenings with advancing age (Feinsilver, 2003; Ohayon, Carskadon, Guilleminault, & Vitiello, 2004). Several cross-sectional studies using self-report measures indicate there are increases in fragmented sleep (Middlekoop et al., 1996; Newman, Enright, Manolio, Haponik, & Wahl, 1997; Schubert et al., 2002) and in difficulty falling asleep with increasing age, particularly in women (Middlekoop et al., 1996; Newman et al., 1997). Other cross-sectional studies indicate there are no overall differences in sleep between young adults and older adults (Gislason et al., 1993). Mood disorders and health problems have been implicated as possible explanations for the conflicting findings (Giron et al., 2002; McCrae et al., 2005; Vitiello et al., 2002).

A strong link exists between poor psychological health and poor sleep (Ohayon & Vecchierini, 2005). Self reported anxiety and worry have been associated with difficulty initiating sleep (Maggi et al., 1998). Depression has been associated with reports of difficulty falling asleep, fragmented sleep, early morning awakening, and waking feeling unrefreshed (Foley et al., 2004; Newman et al., 1997). Poor sleep has been implicated as a risk factor for depression, a consequence of depression, a potential result of medication management for depression, and a common refractory symptom of depression (Benca, 2005).

The relationship between poor physical health and poor sleep has been well studied. Poor self-rated health and the number of health problems are strongly related to poor sleep (Foley et al., 2004). Reports of night time urination, gastrointestinal problems, respiratory disease, and cardiovascular problems have been consistently associated with poor sleep quality as measured by self-report (Taylor et al., 2007). However, when study participants were carefully screened for health conditions, or when health and well-being factors were controlled in analyses, age-related differences were no longer found (Foley et al., 1995; Newman et al., 1997; Vitiello et al., 2002). The general consensus among sleep scientists is that there are age-related changes in sleep-related physiology. However, the negative impact of psychological and physical health conditions on sleep far outweighs that of normal physiologic age-related change (Lesage & Scharf, 2007).

Hamilton and colleagues (2007a) in a study of adults with rheumatoid arthritis and fibromyalgia, posited that sleep buffered the relationship between stress and negative affect, suggesting that while poor sleep may increase the risk of anxiety and depressive symptoms, restorative sleep may serve to manage them. Yet, few studies have examined the potential restorative effects of sleep, including the relationship between sleep and positive psychological functioning. Hamilton and colleagues (2007b) found that adults reporting sleeping 6 – 8.5 hour of sleep nightly had lower levels of anxiety and depression and higher levels of psychological well-being (positive relations with others, purpose in life, and self-acceptance) than those reporting less than 6 hours or more than 8.5 hours per night. Plasma IL-6 levels (known inflammatory biomarkers in older women) were found to be lower in women scoring higher in psychological well-being (i.e., positive relations with others; Friedman, Hayney, Love, Singer, & Ryff, 2005). Lower IL-6 levels predicted greater sleep efficiency, while high plasma IL-6 levels were found in women with lower psychological well-being and reporting lower sleep quality (Friedman et al., 2005).

In summary, there are overall declines in sleep quality with age. There is little consensus about whether a decline in sleep quality is a normative developmental change in old age (Sleep and Aging: National Sleep Disorders Research Plan, 2003). The declines may be due to normal age-related physiologic changes as well increasing psychological and physical health changes. Also poor sleep may contribute to declines in physical and psychological health. Clearly longitudinal investigations to clarify the relationship between sleep quality and health and well-being factors in older adults are needed. Also, previous research has not examined inter- or intra-individual changes in sleep quality with age to determine whether sleep quality varies for subgroups of older adults. Factors that might differentiate these subgroups include psychological distress, psychological well-being, and varying degrees of health. The use of growth models in longitudinal aging research emphasizes inter-individual differences in intra-individual change and does not assume that all individuals follow the same pattern of change over time. The use of growth modeling is expanding exponentially, particularly among those who study psychological factors. This approach is important because the "average" aging adult may not be informative if there is a great deal of variability in health and well-being across adults as they age.

Also neglected in prior research is whether positive psychological functioning may predict changes in sleep quality over time. The literature is clear that anxiety, worry, depression, may be important antecedents of poor sleep, but the role of positive psychological functioning may also be important. Recent research suggests that positive psychological factors may have protective influences. It is unclear whether positive psychological factors may influence declines in sleep quality seen with age. This investigation examines the intraindividual changes in sleep over time and focuses on both the positive and negative predictors of changes in sleep in a unique longitudinal study that allows for assessment of the links between psychosocial factors and longitudinal change in sleep over a decade.

The overall purpose of this investigation was to examine changes in the sleep quality of older women over time and to determine whether psychological well-being (positive relations with others, environmental mastery, personal growth, purpose in life, and self-acceptance), psychological distress (depression and anxiety) or health factors (subjective health and number of illnesses) predict these changes. The specific research questions were:

  1. Does the sleep quality of older women change over time?

  2. Are there different patterns of change in sleep quality over time?

  3. Do psychological well-being (positive relations with others, environmental mastery, personal growth, purpose in life, and self-acceptance), psychological distress (depression and anxiety) or physical health (subjective health, number of illnesses) predict different patterns of change in sleep quality over time?

Method

Design

This was a secondary analysis of data from a longitudinal study of aging women (Kling et al., 1997; Kwan, Love, Ryff, & Essex, 2003). The sample was first recruited for the purpose of studying a major transition of aging (community relocation), then additional funding was obtained to do further longitudinal follow-up of a subsample of respondents for whom contact information was still available.(See Figure 1. Study design.)

Figure 1.

Figure 1

Open in a new tab

Study design.

Sample

Eligibility criteria for the parent studies included English-speaking, community-dwelling women, 55 years or older, with sufficient cognitive and physical health to complete study interviews, questionnaires, and examinations. At baseline, participants were 55 to 84 years old (M = 67, SD =7.2). At 10 years, participants were 63 to 93 years of age (M = 76, SD 7.2).

Differences between completers (N = 115) and non-completers (N = 403) in baseline demographics, sleep quality and health and well-being variables were examined (see Table 1). Completers were younger (t = 4.91, p > .0001), more educated (t = −2.67, p > .01), with higher levels of personal growth (t = −4.02, p > .001), subjective health (t = −3.30, p > .001), and fewer illnesses (t = 3.43, p > .001) suggesting completers may be positively biased. No significant differences were found between groups in sleep quality, anxiety, depression, or other dimensions of psychological well-being.

Table 1.

Comparison of non-completers and completers at baseline

Non-completers (N = 403) Completers (N = 115)
M (SD) M (SD)
Age*** 71.01 (8.15) 66.89 (7.18)
Years of education** 13.48 (2.72) 14.26 (2.93)
Sleep quality 5.12 (3.05) 4.88 (2.73)
Psychological well-being
    Positive relations with others 68.19 (10.14) 68.12 (10.36)
    Autonomy 62.06 (9.41) 62.32 (9.98)
    Environmental mastery 64.38 (9.62) 65.34 (9.50)
    Personal growth*** 66.26 (9.04) 70.17 (9.70)
    Purpose in life 64.23 (9.71) 65.98 (10.30)
    Self-acceptance 62.65 (11.16) 64.02 (11.23)
Psychological distress
    Depression 12.01 (8.45) 10.57 (7.46)
    Anxiety 19.38 (5.63) 19.04 (5.48)
Health
    Subjective health*** 5.11 (1.16) 5.50 (0.92)
    Number of illnesses*** 2.74 (2.0) 2.03 (1.73)
Open in a new tab
*

p < .05.

**

p < .01.

***

p < .001.

Measures

Sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI) (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). The PSQI is a 19-item scale that produces a global sleep quality score; it has demonstrated utility in the assessment of sleep quality and sleep disturbances in older adults (Carpenter & Andryskowski, 1998; Smith, 1999). A global score > 5 yielded a diagnostic sensitivity of 89.6% and specificity of 86.5% (kappa = 0.75, p < 0.001) indicating poor subjective sleep quality (e.g., more sleep disturbances, daytime dysfunction, lower sleep duration, and lower habitual sleep efficiency) (Buysse et al., 1989). The PSQI has overall high internal consistency (Cronbach’s alpha), ranging from .80 to .87 for community samples (Backus, Junghanns, Broocks, Riemann, & Hohagen, 2002) and was .70 in this study. Higher numbers indicate poorer sleep quality.

Psychological well-being was assessed using Ryff’s six scales of psychological well-being Ryff & Keyes, 1995). Positive Relations with Others (PR), emphasizes the achievement of intimacy and indicates satisfying relationships with others. Autonomy (AU) is being self-determined and free of the opinions of others. Environmental Mastery (EM) reflects a sense of competence in managing one’s environment. Personal growth (PG) is viewing oneself as continually growing and changing in ways that reflect self-knowledge. Purpose in Life (PIL) refers to a feeling that life is meaningful. Self-Acceptance (SA) reflects a positive attitude toward oneself and acknowledgment of one’s good and bad qualities. Each scale consists of 14 items with responses ranging from 1 (strongly disagree) to 6 (strongly agree). Higher scores indicate higher psychological well-being. Cronbach’s alphas ranged from .82 to .88.

Psychological distress. The Center for Epidemiological Studies Depression (CES-D) inventory (Radloff, 1977) and the State-Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, & Lushene, 1970) were used to measure psychological distress. The CESD is a 20-item self-report scale designed to measure depressive symptomatology. Respondents were asked to answer each item on a 0 (rare or none) to 3 (most or all of the time) scale, based on how often they felt in the past week. Scores >15 are suggestive of clinical depression. For this study, total scores were adjusted to account for the elimination of a sleep item; the highest possible CES-D score was 57. The Cronbach’s alpha was .83. The 10-item State Anxiety component of the State Trait Anxiety Inventory (STAI; Spielberger, Sydeman, Owen, & Marsh, 1999) measures anxiety-related feelings on an average day). Higher scores reflect greater state anxiety. The alpha coefficient for the STAI was .85.

Physical Health. There were two measures of physical health. Subjective health was assessed with a single-item question, “How would you describe your overall health at the present time?” on a 7-point scale ranging from 1 (poor) to 7 (excellent). Higher scores indicate better health. This measure was selected because self-rated health status has been found to be one of the best predictors of mortality and morbidity among older adults and an important predictor of functional ability and health-related quality of life (Idler & Kasl, 1991; Wilson & Cleary, 1995). Number of illnesses was assessed using the Older Americans Resources Survey (OARS; Duke University, 1978). It is a self-report checklist of illnesses common in midlife and older adults. Participants indicated whether or not they had experienced each of the 19 illnesses listed (e.g., asthma or wheezing, kidney disease, high blood pressure, etc.). The total number of illnesses was summed to yield an illness score. Higher scores indicate poorer health.

Procedure

Women responding to newspaper advertisements, brochures, and housing facility managers invitations were screened via telephone to determine eligibility. Baseline in-home assessments of independence in physical functioning and in-depth personal interviews were conducted by trained interviewers. 518 women completed self-administered questionnaires (SAQ) of psychological well-being at baseline. At the end of the initial investigation, letters were sent to 369 women who had expressed interest in participation in future studies. Participation was contingent on completion of the previous waves of data collection. Women were contacted until a final sample size of 135 was obtained. Reasons for non-participation included: deceased (11.5%), poor health or health of family member (26%), study too demanding (11.5%), unable to contact/out of area (16.3%) and refused for other reasons (34.6%). The 135 women participated in interviews, psychological and laboratory testing including the Mini-mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975), and completed questionnaires during a two-night at the General Clinical Research Center (GCRC) as part of the parent study. Testing was repeated two years later. This investigation includes the 115 women with data at baseline, 8 years, and 10 years (see Figure 1.) The research proposals were approved by the Institutional Review Board (IRB) at the University of Wisconsin-Madison.

Data Analysis

SPSS Version 12.0 (2003) was used for descriptive statistics. Mplus Version 5 (Muthen and Muthen, 2007) was used for growth curve modeling (GCM) and growth mixture modeling (GMM). This approach models a “change trajectory” that reflects the relationship between time and changes in outcome variables within an individual subject. Latent variables representing the intercept (e.g., estimated baseline scores) and slope (e.g., estimated change over time) are then used to model the expectancy of growth across subjects. Individual growth modeling is well-suited for this analysis because older adults are more heterogeneous than alike. Growth mixture modeling allows the examination of subgroups within the larger growth curve model (Wu, 1999).

Results

Sample Characteristic. Participants were well-educated, Caucasian, living alone, and unemployed (Table 2). In general, women had high levels of psychological well-being, low mean levels of depressive symptomatology and anxiety, “good” subjective health and low number of illnesses. The most common illnesses reported at baseline were arthritis, hypertension, glaucoma/cataracts, thyroid problems, and heart disease.

Table 2.

Baseline demographic characteristics of community-dwelling older women (N = 115)

Characteristic n %
Caucasian 112 97
> HS education 40 45
Married or living with partner 32 28
Lived alone 66 57
Employed 45 39
Income < $20K 48 42
Open in a new tab

Does the sleep quality of older women change over time? Overall, sleep quality declined over time from M = 5 (SD = 2.72) to 6.29 (SD = 3.65); nearly half of the women reported the use of sleep aids at Time 3. To identify the general form of change in sleep quality over time, a linear growth curve was tested. Fit statistics indicate that a linear growth curve model was a good fit (X2 = 0.46; p = 0.5; Comparative Fit Index (CFI) = 1; Tucker Lewis Index (TLI) = 1.02; Root Mean Square Error of Approximation (RMSEA) = 0.00, Standardized Room Mean Square Residual (SRMR) = 0.01). The overall growth curve of sleep quality in community-dwelling older women increased significantly over time, indicating reduced sleep quality over time (β0 = 4.97; SE = 0.24; β1 = 0.14; SE = 0.03; p < .001). Note that higher numbers reflect declining sleep quality.

Are there different patterns of change in sleep quality over time? The most parsimonious model (a one-class model) was followed by sequentially increasing the number of growth model classes to three latent class models, with the choice of best fitting model based on the following criteria: the Akaike Information Criterion (AIC) statistic (Akaike, 1974), the Bayesian Information Criterion (BIC), the adjusted BIC statistical fit index (Schwarz, 1978), and the CAIC (Consistent AIC). Lower AIC, BIC, Adjusted BIC, and CAIC values indicate better model fit and a significant Lo-Mendell Rubin Likelihood Ratio (LMR) demonstrating a significant improvement in fit for the inclusion of one more class (Li & Nyholt, 2001). While Adjusted BIC has been found to identify the number of classes better than other fit statistics, it is not perfect. Since no single parameter demonstrates a best fit, examining across the parameters provides the best estimate of fit (Nylund, Muthen, & Asparouhov, 2007).

The three-category model demonstrated the best fit of the three models tested (Table 3). Group 1 (n = 23) had an estimated mean sleep quality score that did not change significantly over time (β0 = 7.34; SD = 0.47; β1 = 0.24; SD = 0.13; p < .058). Group 2 (n = 4) had the worse mean sleep quality score at baseline that became significantly worse over time (β0 = 10.62; SD = 2.74; β1 = 0.61; SD = 0.21; p < .004). Group 3 (n = 88) had the best mean sleep quality score at baseline, but it significantly worsened over time (β0 = 4.06; SD = 0.29; β1 = 0.09; SD = 0.04; p < .009).

Table 3.

Growth Mixture Model Fit Statistics for One-, Two- and Three-Category Sleep Quality Models Original Sample (N=115) and Retested Sample (N = 111)

Adjusted LMR
Model Group N AIC BIC BIC CAIC 2LL p
Original one-category 1 115 1719.69 1741.65 1716.27 1720.93
Original two-category 1 106 1704.83 1735.03 1700.26 1720.93 19.49 0.01
2 9
Original three-category 1 23 1705.64 1744.07 1699.82 1700.06 4.85 0.69
2 4
3 88
Retested one-category 1 111 1614.83 1639.51 1611.22 1626.07
Retested two-category 1 87 1609-34 1639.15 1604.22 1616.07 10.73 0.24
2 24
Retested three-category 1 86 1611.30 1649.23 1604.99 1589.57 3.77 0.14
2 13
3 12
Open in a new tab

Note. AIC = Akaike’s information criterion; BIC = Bayesian information criterion; Adj. BIC = BIC adjusted for sample size; CAIC = Consistent AIC; LMR = Lo-Mendell Rubin Likelihood Ratio.

Preliminary to performing the analysis for question 3, GCM was conducted for each of the health and well-being variables. Fit statistics were examined to ensure that each predictor variable met the assumptions of a linear model. For physical health, number of illnesses was dropped from this step of the analysis because of poor fit (Table 4). For psychological well-being, positive relations, personal growth, and purpose in life did not change, but, autonomy, environmental mastery, and self-acceptance had significant increases. For psychological distress, anxiety and depression scores improved significantly over time. Subjective health declined significantly over time. (See Table 5).

Table 4.

Linear Growth Curve Model Fit Statistics for Each Health and Well-Being Variable (N=115)

Variable X2 p CFI TLI RMSEA SRMR
Psychological well-being
    Positive relations with others 0.65 0.42 1.00 1.00 0.00 0.01
    Autonomy 0.55 0.46 1.00 1.01 0.00 0.01
    Environmental mastery 2.93 0.09 0.99 0.97 0.13 0.03
    Personal growth 0.97 0.33 1.00 1.00 0.00 0.02
    Purpose in life 1.74 0.19 1.00 0.99 0.08 0.02
    Self-acceptance 0.14 0.71 1.00 1.01 0.00 0.01
Psychological distress
    Depression 0.02 0.89 1.00 1.05 0.00 0.003
    Anxiety 1.76 0.18 0.99 0.98 0.08 0.02
Physical health
    Subjective health 2.95 0.09 0.98 0.95 0.13 0.03
    Number of illnesses 7.76 0.005** 0.94 0.83 0.24 0.06
Open in a new tab

Table 5.

Linear Growth Curve Model for Each Health and Well-Being Variable (N = 111)

Variable B0
(Intercept)		 SE B1
(Slope)		 SE
Psychological well-being
    Positive relations with others 68.26 0.96 0.10 0.07
    Autonomy 62.25 0.96 0.18** 0.07
    Environmental mastery 65.64 0.91 0.22* 0.9
    Personal growth 70.28 0.88 −0.06 0.07
    Purpose in life 65.84 0.98 0.06 0.07
    Self-acceptance 64.59 1.04 0.23** 0.80
Psychological distress
    Depression 9.42 0.66 −0.26*** 0.07
    Anxiety 18.92 0.52 −0.32*** 0.05
Subjective health 5.57 0.09 −0.04*** 0.01
Open in a new tab
*

p < .05.

**

p < .01.

***

p < .001.

Do psychological well-being (positive relations with others, environmental mastery, personal growth, purpose in life, and self-acceptance), psychological distress (depression and anxiety) or physical health (subjective health) predict different patterns of change in sleep quality over time? The three-category model could not be used in these analyses because the smallest group (n = 4) was too small for statistical comparisons. The four women in this group were removed from analysis. GCM and GMM modeling were conducted with the remaining 111 women. A sequential testing of one-, two- and three-category models indicated that a two-category model was the best fit, and a better fit than the previous two- and three-category models using all 115 women (see Table 3). In this model, group 1 (n = 87) had relatively good sleep quality scores at baseline, but worsened significantly over time (β0 = 3.95; SE = 0.39; β1 = 0.10; SE = 0.10; p < .01). Group 2 (n = 24) had worse sleep quality scores at baseline that remained poor (β0 = 7.38; SE = 0.42; β1 = 0.20; SE = 0.14; p > .05). A binary classification of sleep quality was created from these distinctly different groups. The larger group (n = 87) had low sleep quality scores at baseline and scores increased significantly over time indicating good but diminished sleep quality over time. This group was classified as the “good sleep” (GS) in that while sleep worsened significantly overtime, overall sleep scores remained within a good vs poor range. The smaller group (n = 24) had poor sleep quality scores at baseline and no change in the level of sleep quality over time, indicating disrupted sleep that persisted over time. This group was classified as “disrupted sleep” (DS).

GMM was conducted to identify whether any psychological well-being, psychological distress, or physical health variables at baseline predicted membership in the GS or DS group. The growth class membership, based on posterior probabilities, was modeled using logistic regression. Predictors were modeled one at a time. The results indicate that baseline levels of five dimensions of psychological well-being (positive relations with others, environmental mastery, personal growth, purpose in life, and self-acceptance) predicted membership in the DS group of women, that is, higher levels of PWB were associated with lower odds of disrupted sleep (see Table 6). Higher depression and more illnesses at baseline were associated with greater odds of having disrupted sleep. Baseline autonomy, anxiety, and subjective health were not significant.

Table 6.

Baseline Health and Well-Being Variables Predicting Membership in the Disrupted Sleep Group (N = 111)

Variable Estimated logit SE Odds ratio
Psychological well-being
    Positive relations with others*** −0.11 0.03 0.90
    Autonomy −0.02 0.03 0.98
    Environmental mastery** −0.07 0.02 0.94
    Personal growth* −0.06 0.02 0.94
    Purpose in life** −0.07 0.02 0.93
    Self-acceptance** −0.07 0.02 0.94
Psychological distress
    Depression* 0.07 0.03 1.08
    Anxiety 0.01 0.04 1.01
Subjective health −0.14 0.25 0.87
Number of illnesses* −0.27 0.12 0.77
Open in a new tab
*

p < .05.

**

p < .01.

***

p < .001.

Discussion

This is the first study to examine individual differences in intraindividual change in sleep among older women, focusing on both positive and negative predictors, in a unique longitudinal study that allows for assessment of links between psychosocial factors and longitudinal change in sleep over a decade. We found a significant decline in the sleep quality of older women over time. Yet not all women experienced the same pattern of decline over time. Three distinctly different patterns of sleep emerged: women with good but declining sleep quality over time, women with poor sleep quality that remained poor over time, and a small number of women with poor sleep that continued to decline over time. While the first group of women had declines in their sleep, their overall sleep quality could still be categorized as good.

In previous studies, depression has been strongly linked to disrupted sleep. In this investigation, depression did not predict membership in the disrupted sleep group. However, baseline depression was a significant predictor of declining sleep quality overall, which suggests that depression has a role in declining sleep quality, but perhaps not for all aging women. For example, it may be beneficial to examine differences between women with and without significant depressive symptoms or with concomitant depression and poor physical health. Anxiety has also been associated with poor quality sleep—in particular, difficulty falling asleep, reduced total sleep time, and difficulty returning to sleep once aroused (Spira, Stone, Beaudreau, Ancoli-Israel, & Yaffe, 2009). Indeed, sleep problems are included in the diagnostic criteria for generalized anxiety disorder, and anxiety is included in the diagnostic criteria for insomnia (American Academy of Sleep Medicine, 2005; American Psychiatric Association, 2000). In this investigation, anxiety was not found to predict changes in sleep quality with aging, and, in fact, an overall decline in anxiety was demonstrated. However, the overall level of anxiety in this sample was low. It is not clear what contributes to either decreases or increases in anxiety with age, and we do not know at what levels of anxiety sleep quality may begin to decline.

The findings specific to PWB suggest that interplay between well-being and sleep exists (Ryff, Singer, & Love, 2004). Previous research in aging has focused on the negative aspects of psychosocial health—that is, how psychological adversity increases biological risk. While this is an important area of investigation, it ignores the potential benefits of positive health. A key hypothesis of positive health is that there is interplay between biopsychosocial processes such that positive psychological well-being will be accompanied by positively functioning physiologic processes. This interplay has been demonstrated in studies of PWB and health, that is lower plasma IL-6 levels (Friedman, Hayney, Love, Singer, & Ryff, 2005) and lower glycosylated hemoglobin (Tsenkova, Love, Singer, Ryff, 2007) in individuals with higher PWB. If the PWB findings of this investigation were interpreted from a positive health perspective, findings from this investigation indicate that higher levels of baseline positive relations with others, environmental mastery, purpose in life, and self-acceptance predicted membership in the GS class. The nature of the relationships between PWB and sleep needs further examination in larger, longitudinal studies, and in more heterogeneous samples.

It has been assumed that older adults follow the same pattern of sleep with age, except for variances due to specific health problems. In this study, we examined two measures of physical health in relation to sleep quality. A decline in subjective health did not predict declines in the sleep quality of older women. Because of sample size limitations, we were not able to examine whether health at baseline or declines in health over time predict changes in sleep quality for some groups and not others. The small group of women with very poor sleep quality may represent a group of women with poorer health and investigation of this group is indicated. It would be interesting to examine whether positive health factors, such as exercise and/or lower body mass index, might protect women from disrupted sleep over time. Exercise, cognitive behavioral therapies, and weight reduction are interventions shown to improve disrupted sleep. It is not clear whether use of these techniques prior to declines in sleep quality might serve a protective role. Another ideal question for future inquiry is whether restorative sleep may serve to buffer the relationship between stress and negative affect.

The women in this sample were generally well-functioning, healthy older women and still experienced declines in sleep quality. This seems to support the view that the shift toward lower quality sleep with aging is normal. On the other hand, if diminishing health and well-being are largely responsible for the changes in sleep observed with age, it would seem likely that a group of healthy women with consistently high sleep quality would emerge. When conducting a meta-analysis of age-related changes in sleep, Ohayon et al., (2004) found that the effect size was considerably smaller when participants with psychological and/or physical illness were excluded, but that poor health did not obscure the overall significant declines in sleep quality with age. These findings are consistent with this study, which suggests that age-related changes in sleep quality are normative and are independent of diminished health and well-being. What is not known is whether the findings of this investigation reflect changes in sleep in aging men or even in women who are not so healthy or socieoeconomically advantaged.

In this investigation, statistical methods were employed that have not been used in past sleep quality research; consequently, they provide another perspective for examining potential changes in sleep quality over time. The use of GCM and GMM allowed the sleep quality of individual women to be followed over time and their rates of change in sleep quality to be analyzed. Following intra-individual and inter-individual change over time provides insights into why previous research findings regarding sleep changes with age were conflicting. Also, this study used a community-based sample of older adults. Research on community groups provides a different picture than research on clinical samples, expanding previous sleep research of healthy older adults. Finally, few studies have examined the relationship between sleep and positive psychological functioning. The findings of this investigation suggest that PWB may serve as a resource that deserves more attention.

Limitations

As a secondary data analysis, the design and measures used in the study could not be changed. Having data at four intervals, rather than three may have allowed for testing for curvilinear changes over time. This study relied on self report measures (PSQI), which has limitations. There is a tendency for some subjects to underestimate sleep time because of fragmented sleep, a problem for many older adults. However, study of sleep disruption often relies on self-report because the cost of PSG can be prohibitive for research not involving sleep staging or the diagnosis and management of sleep disorders. Also, PSG findings have been found to be at variance with subjective sleep complaints (Pressman & Fry, 1988). Taking the limitations of both methods into consideration, self-report findings must be interpreted with caution. Future studies may benefit from combining self report with polysomnography measures.

We were able to examine predictors of different sleep patterns, but we were not able to test the relative contribution of different dimensions of PWB to sleep quality because of sample size limitations. We also did not examine the potential bidirectional influence of disrupted sleep on health and well-being. Both would be important directions for future studies. And finally, the sample for this investigation was skewed toward positive health and well-being. Women had few illnesses, were economically secure, and were mainly Caucasian. Cultural differences in sleep quality have been reported in a multiethnic sample of middle-aged women (Jean-Louis, Maqua et al., 2008). Conducting similar analyses with a more culturally or economically heterogeneous group and in women with poorer health is important as it may produce different outcomes. However, in this sample of relatively healthy women, higher levels of several dimensions of PWB at baseline predicted reduced odds for DS group membership, perhaps protecting women from disrupted sleep over time.

Acknowledgments

This research was supported in part by the Helen Denne Schulte fund, School of Nursing, University of Wisconsin–Madison (Dr. Phelan) and by a University of Wisconsin–Madison School of Nursing Research Committee Award (Dr. Heidrich), NIA Grant R01–AG0879 and NIMH Grant P50–MH61083 (Dr. Ryff), NIH Grant M01–RR03186 (UW General Clinical Research Center).

Footnotes

Publisher's Disclaimer: The following manuscript is the final accepted manuscript. It has not been subjected to the final copyediting, fact-checking, and proofreading required for formal publication. It is not the definitive, publisher-authenticated version. The American Psychological Association and its Council of Editors disclaim any responsibility or liabilities for errors or omissions of this manuscript version, any version derived from this manuscript by NIH, or other third parties. The published version is available at www.apa.org/pubs/journals/pag

Contributor Information

Cynthia H. Phelan, William S. Middleton Memorial VA Geriatric Research Education and Clinical Center

Gayle D. Love, University of Wisconsin–Madison Institute on Aging

Carol Ryff, University of Wisconsin–Madison Institute on Aging.

Roger L. Brown, University of Wisconsin–Madison School of Nursing

Susan M. Heidrich, University of Wisconsin–Madison School of Nursing

References

  1. Akaike H. A new look at the statistical model identification. IEEE Transactions on Automatic Control. 1974;19:716–723. [Google Scholar]
  2. American Academy of Sleep Medicine. International classification of sleep disorders: Diagnostic and coding manual. 2nd ed. Westchester, IL: American Academy of Sleep Medicine; 2005. [Google Scholar]
  3. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (4th ed., text revision; DSM-IV-TR) Arlington VA: 2000. [Google Scholar]
  4. Ancoli-Israel S. Sleep and its disorders in aging populations. Sleep Medicine. 2009;10:S7–S11. doi: 10.1016/j.sleep.2009.07.004. [DOI] [PubMed] [Google Scholar]
  5. Backus J, Junghanns K, Broocks A, Riemann D, Hohagen F. Test-retest reliability and validity of the Pittsburgh sleep quality index in primary insomnia. Journal of Psychosomatic Research. 2002;53:737–740. doi: 10.1016/s0022-3999(02)00330-6. [DOI] [PubMed] [Google Scholar]
  6. Benca R. Mood disorders. In: Kryger MH, Roth T, Dement WC, editors. Principles and practice of sleep medicine. 4th ed. Baltimore: Elsevier Saunders; 2005. pp. 1311–1326. [Google Scholar]
  7. Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh sleep quality index: A new instrument for psychiatric practice and research. Psychiatry Research. 1989;28:193–213. doi: 10.1016/0165-1781(89)90047-4. [DOI] [PubMed] [Google Scholar]
  8. Carpenter JS, Andrykowski MA. Psychometric evaluation of the Pittsburgh sleep quality index. Journal of Psychosomatic Research. 1998;45:5–13. doi: 10.1016/s0022-3999(97)00298-5. [DOI] [PubMed] [Google Scholar]
  9. Duke University Center for the Study of Aging and Human Development. Multidimensional functional assessment: The OARS methodology. Durham, NC: Duke University; 1978. [Google Scholar]
  10. Feinsilver SH. Sleep in the elderly: What is normal? Clinics of Geriatric Medicine. 2003;19:177–188. doi: 10.1016/s0749-0690(02)00064-2. [DOI] [PubMed] [Google Scholar]
  11. Floyd JA. Sleep and aging. Nursing Clinics of North America. 2002;37:719–731. doi: 10.1016/s0029-6465(02)00034-8. [DOI] [PubMed] [Google Scholar]
  12. Foley D, Ancoli-Israel S, Britz P, Walsh J. Sleep disturbances and chronic disease in older adults: Results of the 2003 National Sleep Foundation Sleep in America Survey. Journal of Psychosomatic Research. 2004;56:497–502. doi: 10.1016/j.jpsychores.2004.02.010. [DOI] [PubMed] [Google Scholar]
  13. Foley DJ, Monjan A, Brown SL, Simonsick EM, Wallace RB, Blazer DG. Sleep complaints among older persons: An epidemiologic study of three communities. Sleep. 1995;18:425–432. doi: 10.1093/sleep/18.6.425. [DOI] [PubMed] [Google Scholar]
  14. Folstein MF, Folstein SE, McHugh PR. Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research. 1975;12:189–198. doi: 10.1016/0022-3956(75)90026-6. [DOI] [PubMed] [Google Scholar]
  15. Friedman EM, Hayney MS, Love GD, Urry HL, Rosenkranz MA, Davidson RJ, et al. Social relationships, sleep quality, and interleukin-6 in aging women. Proceedings of the National Academy of Science of the United States of America. 2005;1002:18757–18762. doi: 10.1073/pnas.0509281102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ganguli M, Reynolds CF, Gibley JE. Prevalence and persistence of sleep complaints in a rural older community sample: The MoVIES project. The Journal of the American Geriatrics Society. 1996;44:778–784. doi: 10.1111/j.1532-5415.1996.tb03733.x. [DOI] [PubMed] [Google Scholar]
  17. Giron MST, Forsell Y, Bernsten C, Thorslund M, Windblad B, Fastbom J. Sleep problems in a very old population: Drug use and clinical correlates. Journal of Gerontology. 2002;37A:M236–M240. doi: 10.1093/gerona/57.4.m236. [DOI] [PubMed] [Google Scholar]
  18. Gislason T, Reynisdottir H, Kristbjarnarson H, Benediktsdottir B. Sleep habits and sleep disturbances among the elderly: An epidemiologic study. Journal of Internal Medicine. 1993;234:31–39. doi: 10.1111/j.1365-2796.1993.tb00701.x. [DOI] [PubMed] [Google Scholar]
  19. Hamilton NA, Catly D, Karlson C. Sleep and the affective response to stress and pain. Health Psychology. 2007a;26:288–295. doi: 10.1037/0278-6133.26.3.288. [DOI] [PubMed] [Google Scholar]
  20. Hamilton NA, Nelson CA, Stevens N, Kitzman H. Sleep and psychological well-being. Social Indicators Research. 2007b;82:147–163. [Google Scholar]
  21. Hu L, Bentler PM. Cutoff criteria for fit indices in covariance structure analysis: Conventional criteria vs. new alternatives. Structural Equation Modeling. 1999;6:1–55. [Google Scholar]
  22. Idler EL, Kasl S. Health perceptions and survival: Do global evaluations of health status really predict mortality? Journals of Gerontology: Social Sciences. 1991;46:55–65. doi: 10.1093/geronj/46.2.s55. [DOI] [PubMed] [Google Scholar]
  23. Li W, Nyholt DR. Marker selection by Akaike information criterion and Bayesian information criterion. Genetic Epidemiology. 2001;21(supp 1):S272–S277. doi: 10.1002/gepi.2001.21.s1.s272. [DOI] [PubMed] [Google Scholar]
  24. Jacobs JM, Hammerman-Rozenberg R, Cohen A, Stessman J. Chronic back pain among the elderly: Prevalence, associations, and predictors. Spine. 2006;31:E203–E207. doi: 10.1097/01.brs.0000206367.57918.3c. [DOI] [PubMed] [Google Scholar]
  25. Jean-Louis G, et al. Insomnia symptoms in a multiethnic sample of American Women. Journal of Women’s Health. 2008;17:15–25. doi: 10.1089/jwh.2006.0310. [DOI] [PubMed] [Google Scholar]
  26. Kling KC, Seltzer MM, Ryff CD. Distinctive late-life challenges: Implications for coping and well-being. Psychology and Aging. 1997;12:288–295. doi: 10.1037//0882-7974.12.2.288. [DOI] [PubMed] [Google Scholar]
  27. Kwan CM, Love GD, Ryff CD, Essex MJ. The role of self-enhancing evaluations in a successful life transition. Psychology of Aging. 2003;18:3–12. doi: 10.1037/0882-7974.18.1.3. [DOI] [PubMed] [Google Scholar]
  28. Lesage S, Scharf SM. Beyond the usual suspects: Approaching sleep in elderly people. Journal of Gerontology. 2007;62A:53–54. doi: 10.1093/gerona/62.1.53. [DOI] [PubMed] [Google Scholar]
  29. Littner M, Hirshkowitz M, Kramer M, Kapen S, Anderson W, Bailey D, et al. Practice parameters for using polysomnography to evaluate insomnia: An update. Sleep. 2003;26:754–760. doi: 10.1093/sleep/26.6.754. [DOI] [PubMed] [Google Scholar]
  30. Maggi S, Langlois JA, Minicuci N, Grigoletto F, Pavan M, Foley DJ, et al. Sleep complaints in community-dwelling older persons: Prevalence, associated factors, and reported causes. Journal of the American Geriatric Society. 1998;46:161–168. doi: 10.1111/j.1532-5415.1998.tb02533.x. [DOI] [PubMed] [Google Scholar]
  31. McCrae CS, Rowe MA, Tierney CG, Dautovich ND, DeFinis AL, McNamara JP. Sleep complaints, subjective and objective sleep patterns, health, psychological adjustment, and daytime functioning in community-dwelling older adults. Journal of Gerontology. 2005;60B:P182–P198. doi: 10.1093/geronb/60.4.p182. [DOI] [PubMed] [Google Scholar]
  32. Middlekoop HA, Smilde-van den Doel DA, Neven AK, Kamphuisen HA, Springer CP. Subjective sleep characteristics of 1485 males and females aged 50–93: Effects of sex and age, and factors related to self-evaluated quality of sleep. The Journal of Gerontology. 1996;51A:M108–M115. doi: 10.1093/gerona/51a.3.m108. [DOI] [PubMed] [Google Scholar]
  33. Muthen LK, Muthen BO. Mplus User’s Guide. Fifth Edition. Los Angeles CA: Muthen & Muthen; 1998–2007. [Google Scholar]
  34. National Heart, Lung, and Blood Institute. 2003 National Sleep Disorders Research Plan. U.S. Department of Health and Human Services. 2003 Jul; (NIH Publication No. 03-5209). [Google Scholar]
  35. Newman AB, Enright PL, Manolio TA, Haponik EF, Wahl PW. Sleep disturbance, psychosocial correlates, and cardiovascular disease in 5201 older adults: The cardiovascular health study. Journal of the American Geriatrics Society. 1997;45:1–7. doi: 10.1111/j.1532-5415.1997.tb00970.x. [DOI] [PubMed] [Google Scholar]
  36. Nylund KL, Muthen BO, Asparouhov T. Deciding on the number of classes in latent class analysis: A Monte Carlo simulation study. Structural Equation Modeling: An Interdisciplinary Journal. 2007;14:535–569. [Google Scholar]
  37. Ohayon MM, Carskadon MA, Guilleminault C, Vitiello MV. Meta-analysis of quantitative sleep parameters from childhood to old age in healthy individuals: Developing normative sleep values across the human lifespan. Sleep. 2004;27:1255–1273. doi: 10.1093/sleep/27.7.1255. [DOI] [PubMed] [Google Scholar]
  38. Ohayon MM, Vecchierini M. Normative sleep data, cognitive function, and daily living activities in older adults in the community. Sleep. 2005;28:981–989. [PubMed] [Google Scholar]
  39. Pressman MR, Fry JM. What is normal sleep in the elderly? Clinics of Geriatric Medicine. 1988;4:71–81. [PubMed] [Google Scholar]
  40. Radloff LS. The CES-D scale. A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401. [Google Scholar]
  41. Ryff CD, Singer BH, Love G. Positive health: connecting well-being with biology. Philosophical Transactions of the Royal Society of London Series B Biological Sciences. 2004;359:1383–1394. doi: 10.1098/rstb.2004.1521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Ryff CD, Keyes CLM. The structure of psychological well-being revisited. Journal of Personality and Social Psychology. 1995;69:719–727. doi: 10.1037//0022-3514.69.4.719. [DOI] [PubMed] [Google Scholar]
  43. Schubert CR, Cruickshanks KJ, Dalton DS, Klein BEK, Klein R, Nondahl DM. Prevalence of sleep problems and QOL in an older population. Sleep. 2002;25:889–893. [PubMed] [Google Scholar]
  44. Spielberger CD, Gorsuch RL, Lushene RE. STAI Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press; 1970. [Google Scholar]
  45. Spielberger CD, Sydeman SJ, Owen AE, Marsh BJ. Measuring anxiety and anger with the State-Trait Anxiety Inventory (STAI) and the State-Trait Anger Expression Inventory (STAXI) In: Maruish Mark E., editor. The use of psychological testing for treatment planning and outcomes assessment. 2nd ed. Mahwah, NJ: Lawrence Erlbaum Associates; 1999. pp. 993–1021. [Google Scholar]
  46. Spira AP, Stone K, Beaudreau SA, Ancoli-Israel S, Yaffe K. Anxiety symptoms and objectively measured sleep quality in older women. American Journal of Geriatric Psychiatry. 2009;17:136–143. doi: 10.1097/JGP.0b013e3181871345. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. SPSS for Windows. Chicago: SPSS Inc.; 2003. Sep 4, [Google Scholar]
  48. Taylor DJ, Mallory LJ, Lichstein KL, Durrence HH, Riedel BW, Bush AJ. Comorbidity of chronic insomnia with medical problems. Sleep. 2007;30:213–218. doi: 10.1093/sleep/30.2.213. [DOI] [PubMed] [Google Scholar]
  49. Tsenkova VK, Love GD, Singer BH, Ryff CD. Socioeconomic status and psychological well-being predict cross-time change in glycosylated hemoglobin in older women with diabetes. Psychosomatic Medicine. 2007;69:777–784. doi: 10.1097/PSY.0b013e318157466f. [DOI] [PubMed] [Google Scholar]
  50. Vitiello MV, Moe KE, Prinz PN. Sleep complaints cosegregate with illness in older adults: clinical research informed by and informing epidemiological studies of sleep. Journal of Psychosomatic Research. 2002;53:555–559. doi: 10.1016/s0022-3999(02)00435-x. [DOI] [PubMed] [Google Scholar]
  51. Wu YB, Clopper RR, Woolridge PJ. A comparison of traditional approaches to hierarchical linear modeling when analyzing longitudinal data. Research in Nursing & Health. 1999;22:421–434. doi: 10.1002/(sici)1098-240x(199910)22:5<421::aid-nur8>3.0.co;2-q. [DOI] [PubMed] [Google Scholar]

RESOURCES